PROJECT SUMMARY Obesity has emerged as a potential threat to future increases in life expectancy, but the extent of this threat is uncertain. A key uncertainty is how the obesity-mortality changes over the life course. Current studies have yielded mixed findings. As most deaths in the United States occur at older ages, the age dependency of the obesity-mortality association has important ramifications for estimates of the number of total deaths attributable to obesity. The overall objective of this project is to study the pattern of obesity's effect on mortality over the life course and investigate the effects of selection biases and population heterogeneity with regard to duration of exposure to obesity on this pattern. Moreover, I intend to capture the life-long body mass index trajectories, estimate their mortality risks, and study how they may contribute to changing health disparities over the life course among various sociodemographic groups. The data come from Framingham Heart Study data. I will use both the original Framingham generation (n=1,877) and the offspring cohort (n=2,859). More specifically, this proposal intends to reach three aims. Aim 1: Investigate the changing effect of obesity on mortality over the life course among two birth cohorts, examining (1) whether and how the healthy participant effect and mor- tality selection bias contribute to this changing effect, and (2) whether and how these contributions change across birth cohorts. Aim 2: Estimate latent body mass index (BMI) trajectories over the life course within each birth cohort and investigate whether and how these trajectories are associated with mortality. This analysis will shed light on how population heterogeneity with regard to duration of exposure to obesity may contribute to the age pattern of obesity's effect over life course. Moreover, examination of life-long BMI trajectories will give a much more complete picture of their consequences for mortality than prior studies. Aim 3: Examine whether and how body mass index trajectories contribute to the difference in life expectancy and evolving health dispar- ities across life course among different sociodemographic groups (gender and education). This project is inno- vative because it proposes an unprecedented, longitudinal cohort study design and uses advanced economet- rics methods to detect the contributions of healthy participants, mortality selection and timing of initiation and duration of obesity to the changing obesity's effect on mortality over life course. It will employ relatively com- plete cohort data to estimate life-long body mass index trajectories and their associated mortality risks which should provide a much more accurate estimate of the mortality consequence of obesity than prior studies. Moreover, it will investigate the effect of heterogeneity in body mass index trajectories on difference in life ex- pectancy and the changing health disparities over life course across various sociodemographic groups which have not been sufficiently studied in current epidemiology and medical research. Results from this project will have important implication on the life course pattern of obesity's effect on mortality, the overall consequence of obesity on mortality, and the effect of obesity on group difference in health over life course and life expectancy.